JPH0429012B2 - - Google Patents

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention is directed to a first and a second object having a polymerization relationship in close proximity to each other, in an in-plane direction perpendicular to the polymerization direction. The present invention relates to a device for measuring the amount of relative displacement between objects using diffraction of coherent waves and interference phenomena, and particularly relates to an improvement that enables the amount of relative displacement on the order of minute dimensions to be measured with high precision.

[Conventional technology]

There have been many different methods for measuring the relative displacement between two objects stacked vertically or horizontally, but it is possible to measure the relative displacement between two objects stacked vertically or horizontally. There are surprisingly few devices that aim to measure displacement with high precision.

In particular, not only is it not possible to directly project a light source or lens onto either of the two objects, but the superposition distance between the two objects is extremely close to each other, with the distance between them being on the order of several tens of micrometers. Under conditions where it is not possible to insert any optical components, including lenses and apertures, the only conventional method that seems to be usable is the method described by DCFlanders et al.
Publicly known literature: There was only one method published in Appl. phys. Lett.; 31 (1977) 426.

As the principle of this method is shown in Figure 5,
When the first object o1 and the second object o2 are stacked apart from each other by a small distance z (shown here as vertically stacked vertically), each object o1, o2 has a diffraction grating D _a , D _b
By providing only
The purpose is to measure the relative displacement of both objects o1 and o2 in the in-plane direction x perpendicular to the superimposition direction by using the interference phenomenon, and these two diffraction gratings D _a ,
The pitch or lattice constant d of D _b is the same, and the width in the in-plane direction is also about the same.

In order to measure the amount of relative displacement with such a structure, either the first or second diffraction grating D _a ,
In principle, a coherent wave _i is incident on D _b from the outside at an incident angle of zero (that is, along the normal direction). A typical example of coherent waves is coherent light. In the case of Figure 5, in the figure,
The diffraction grating D _a shown below is shown as having coherent light I _i incident on it from below, so if you follow this, the incident light I _i will be the first stage from the perspective of the incident light. The first diffraction grating D _a generates a plurality of outgoing light groups in respective diffraction angle directions determined according to the order of the first diffraction grating D a .

For simplicity, the figure shows the above-mentioned lattice constant d, incident light wavelength λ, and order “1” in the normal direction.
Only the first-order diffracted lights _a ⁺ and _a ^- emitted in the direction of the diffraction angle _± θ determined according to In fact, as mentioned above, higher-order diffracted light is also included. For convenience, clockwise rotation on the drawing with respect to the normal direction is expressed as positive direction "+",
Set left rotation as negative direction “-”.

The respective emitted light components of the first stage or first diffraction grating D _a are then incident on the second diffraction grating D _b .

As a result, a similar diffraction phenomenon occurs for this second diffraction grating D _b , and therefore, as before, the incident light _a ⁺ , _a ^- , given from the first diffraction grating D _a
Regarding _a o, even if the diagram is limited to the first-order and zero-order diffracted lights that diffracted it, the +θ direction is only affected by the first diffraction grating with respect to the normal direction of the second diffraction grating D _b . Diffracted outgoing light _a ⁺ , second diffraction grating
Outgoing light _b ⁺ is diffracted only by D _b , and outgoing light _{a, b} ⁺ is sequentially diffracted by both diffraction gratings D _a and D _b . In the direction −θ, the outgoing light a − is diffracted only by the first diffraction grating D _a , the outgoing light _b ⁻ is diffracted only by the second diffraction grating D _b , and the outgoing light _b ⁻ is diffracted only by the second diffraction grating D b .
A group of emitted lights _a,b ^- is generated which are sequentially diffracted by D _a and D _b .

Therefore, in the conventional method mentioned here, ±θ is applied to the normal direction of the second diffraction grating D _b as described above.
The output light component groups emitted in two directions defined by are collectively defined as positive direction output light I _p ⁺ and negative direction output light _p ^- respectively, and each is captured by an appropriate photodetector (not shown) and both are detected. intensity difference (I _p ⁺ − _p ⁻ )
Based on the change, both objects are
The relative displacement of o1 and o2 was measured.

[Problem that the invention seeks to solve]

However, in the conventional method according to the above-mentioned known literature,
The change in the light intensity difference (I _p ^{+ -I p -) between the positive and negative output lights I p +} _{and I p} ^- _with ^respect _to ^the diffraction angle θ is not only due to the relative displacement in the x direction (in-plane direction) but also due to the There is a drawback that it depends greatly on the distance difference z between o1 and o2.
For example, in Fig. 5, this means that the diffracted beams _a ⁺ and _b ⁺
As can be seen from the comparison, light components having optical path differences based on the distance z between the objects are mixed into the components of the output light I _p ⁺ and I _p ^- in the directions of the positive and negative diffraction angles. It's for a reason.

In addition, as mentioned above, the positive and negative output lights I _p ⁺ and I _p ^- with respect to the diffraction angle θ also in principle contain higher-order diffracted light components, and furthermore, in Fig. 5, the imaginary line indicates As shown, the diffraction component _r that was reflected and diffracted by the second diffraction grating D _b enters the first diffraction grating D _a again, where it is reflected and diffracted again and returns to the second diffraction grating D _b . After repetition, a multiple reflection mechanism occurs in which the light is mixed into both the positive and negative output light components, making it impossible to avoid the effects of such complex component relationships.
Therefore, together with the fact that the above-mentioned optical path difference exists, the output light intensity difference (I _p ⁺ -I _p ^- ) and the amount of relative displacement in the x direction not only include the distance z between the objects as a parameter, but also This results in an extremely complex functional relationship.

Therefore, even if this functional form is analyzable for a certain distance z, the distance z
If there is even a slight change in the value, it becomes meaningless; for example, if the first object o1 is a semiconductor substrate or semiconductor wafer, and the second object o2 is attached to the surface of the substrate to construct an electronic circuit, etc. In the case of lithography masks that are often removed or replaced, it may be easy to individually form diffraction gratings D _b with the same characteristics on each type of mask, but it is difficult to use them before using them. In this case, it becomes necessary to strictly position the substrate o1 in the direction of polymerization. With this, there is no hope that it will be put into practical use, and there are no reports that it has actually been put into practical use.

Particularly, as mentioned above, the phenomenon of multiple reflections and repeated back and forth movements based on the reflected diffraction component _r indicated by the imaginary line is a rather serious problem. For example, even if it were possible to suppress the dimensional variation in the direction of superposition by placing both objects o1 and o2 extremely close together, repeating multiple reflections would amplify this variation. It is from.

Furthermore, since the final output lights I _p ⁺ and I _p ^- must be detected by separate light intensity detectors, for example, as mentioned above, if the object o2 is a semiconductor substrate or a semiconductor wafer, If o1 is the lithographic mask, the detection system occupies an area of space above the lithographic mask o1, which is undesirable. As shown in , it is not possible to separately arrange the light source and the detector on the upper and lower outer sides of the two objects o1 and o2. This is because the semiconductor substrate or semiconductor wafer o2 is generally placed on a holder that physically supports it.

The present invention was developed in view of these points, and is applicable to situations where two objects whose relative displacements are to be measured are extremely close to each other, and no optical components can be inserted between them. Moreover, even if a condition is imposed in which a light source or a wave intensity detector can only be placed outside of one of these two objects in the direction of superposition, it will not have the drawbacks detailed in the above-mentioned known literature. It is an object of the present invention to provide a device which has a simple configuration, has no dependence on the distance in the direction of superposition between two objects in principle, and is capable of measuring the amount of relative displacement with high precision.

[Means for solving problems]

The disadvantage of the conventional example according to the above-mentioned known documents is that each of the output lights I _p ^{+ and} I _p ^- , which are ultimately the targets of detection, contains components that have followed optical paths of different lengths, especially the Multiple reflections between the first and second diffraction gratings,
All of the multiple waves that have traveled back and forth are superimposed on the final output wave, and this creates the problem of optical path differences for each of the multiple diffraction order components ranging from the 0th order to the mth order (m is an integer). This is due to the fact that it is easy to use.

On the other hand, if we were to extract and select in advance only the waves of a specific order from among all the orders of diffracted light components emitted from the first diffraction grating D _a , using some suitable optical system. If this can then be applied to the second diffraction grating D _b , the above disadvantages will be alleviated. In fact, when two objects to be measured for relative displacement are quite far apart in the direction of superimposition, the first and second objects are By interposing an optical system for selecting and extracting a specific order between two objects, it is possible to adopt a method of eliminating the influence of other order components before making the light incident on the second diffraction grating.

However, in cases where the first and second objects are extremely close to each other, as is the object of the known literature and the present invention, as described above, there is no need to select or extract a specific order between them. Not only is it not possible to arrange the It is still difficult to select and extract a specific order of the diffracted light output from the diffraction grating.

This is because, as in the relationship between a semiconductor substrate or a semiconductor wafer and its lithography mask mentioned above, when the distance between them in the superposition direction is at most several tens of μm, the first The output wave that leaves the diffraction grating may reach the second diffraction grating before it changes from a spherical wave to a plane wave, and if this is the case, this itself is clear from the known theory. This is because it is nearly impossible to geometrically define the diffracted waves of a specific order, no matter what kind of optical system is used within the practical range.

Therefore, in order to achieve the above object, the present invention has the following objectives:
If the second object is so close that an optical component cannot be placed between them, or even if it is placed, it is useless for extracting a specific order, that is, the diffracted light output from the first diffraction grating Even if the wave is incident on the second diffraction grating provided on the second object before it becomes a plane wave, as a result, the wave intensity detector will experience an adverse effect due to the change in the distance in the direction of superimposition. The first configuration is one that can avoid the influence of multiple reflections between the first and second diffraction gratings, and can perform highly accurate measurements of the amount of relative displacement in the in-plane direction. A first diffraction grating provided on an object is irradiated with a coherent wave over a predetermined width, but first and second effective diffraction portions are separated from each other along the in-plane direction. The second diffraction grating provided on the second object is narrower than the first diffraction grating and has first and second effective diffraction portions of the first diffraction grating when viewed from a plane projection. The wave intensity detector is formed by a second diffraction grating to occupy all or at least a part of the area on the second object that overlaps with the spaced area between them, including the center in the in-plane direction. It is proposed that the diffraction grating be provided at a position to receive the wave that is diffracted and reflected and then outputted through a spaced region between the first and second effective diffraction portions of the first diffraction grating.

[Effect]

According to the device of the present invention having the above configuration, for the second diffraction grating positioned according to the above configuration requirements,
The wave that passes through the second effective diffraction portion of the first diffraction grating according to the above configuration requirements, is diffracted by them, and enters the wave, regardless of its order, that is,
Even if it is not selected only for a specific order (actually, as mentioned above, even if you want to select it, it is almost impossible), it is possible for each order to be symmetrically incident on the second diffraction grating at that phase. can. Since this symmetry is maintained even if the distance between the first and second objects changes in the direction of overlapping, the wave intensity detector will eventually detect the change in the direction of overlap as compared to the change in the in-plane direction. It does not affect the detected intensity.

In particular, as long as the above-mentioned structural requirements of the present invention are satisfied, by taking into consideration various design dimensional conditions, the wave components that are repeatedly reflected and diffracted between the first and second diffraction gratings are input to the wave intensity detector. You can prevent it from happening. This is very different from the case according to the known literature, where no matter how the dimensional conditions and the like are designed, the incorporation of multiple reflections and diffraction components cannot be prevented in principle.

Further, as recognized in the above configuration requirements, a single wave intensity detector may be used, and the wave intensity detector can be placed on the same side as the first diffraction grating irradiated with the incident wave. Therefore, for example, since the second object on which the second diffraction grating is provided is a semiconductor substrate or a semiconductor wafer, and this is supported by a suitable holder, there is nothing on the back side of the substrate or wafer. The present invention can be effectively used even in cases where even an optical system cannot be arranged. In principle, the number of wave intensity detectors may be just one, which makes it possible to obtain a measuring device that is simpler and occupies a smaller area for the detection system.

Note that in the device of the present invention, although no order selection means is used between the first diffraction grating and the second diffraction grating, the first and second effective diffraction portions of the first diffraction grating and the second By appropriately selecting the arrangement relationship with the diffraction grating, the dimensional relationship, and each grating constant, etc., the result is that undesirable order wave components among the multiple orders of diffraction wave components diffracted by each diffraction grating. can be excluded from the waves finally given to the wave intensity detector, but in some cases, the incident wave may contain components that have been diffracted by each diffraction grating at the zero order, or components that have passed through the object as they are. There is no particular hindrance to superimposing such a component as described above, that is, a transmitted component that can essentially be said to be the incident wave itself, on the final signal wave.

Even in such a case, if the distance between the objects is not that far, there may be no problem, and existing methods such as detecting the incident wave intensity separately and subtracting it while matching the phase may be used. If a separate detection system capable of electrically discriminating the superimposed components is provided using the technique described above, the apparatus of the present invention can be used effectively as is.

However, in special cases, problems may occur if the intensity of this transmitted component is too large and completely masks the signal wave, or if phase modulation is applied to the signal wave. . In such a case, of course, it is desirable to suppress the generation of the superimposed component or remove the generated one, but as will be seen in the later examples, this is the first and second effective method. Only in the diffraction part,
If we adopt a method of applying incident waves locally or irradiating the entire surface of the first diffraction grating while covering the area between them with an appropriate mask, it is natural that the first,
Wave components that are transmitted to the next stage by zero-order diffraction through the portion between the second effective diffraction portions can be prevented from being generated from the beginning.

In any case, as seen in the device disclosed in the known document described with reference to FIG. Although the incidence of higher-order diffraction waves and reflected diffraction waves cannot be avoided even with good design, if the first diffraction grating is a split type diffraction grating consisting of first and second effective diffraction parts as in the present invention, By simply designing the arrangement, width dimensions, and grating constants of this segmented diffraction grating and the non-segmented second diffraction grating that faces it, these higher-order diffraction wave components and noise components due to reflected diffraction can be effectively eliminated. can be eliminated or reduced.

Further, the method itself of dividing the diffraction grating into first and second effective diffraction parts is not directly defined by the present invention, and any known existing technology may be used.
Simply put, as mentioned earlier, it is possible to locally apply incident waves only to a specific part of a normal diffraction grating, or conversely, to plan for full-surface irradiation and cover the area other than the specific part with a mask. Alternatively, by individually cutting diffraction grating grooves in two regions spaced apart in the in-plane direction, the first and second effective diffraction portions may be completely separated from each other in terms of structure. In the case of structural separation, the wave that ultimately passes through the gap is received by the wave detector, so
The problem of reflection is further reduced, and the reduction in the intensity of the wave can also be suppressed, which is advantageous in improving the signal-to-noise ratio (S/N).

Further, in the above summary structure, as long as the wave is coherent, its name or frequency does not matter. Therefore, although waves with wavelengths in the optical region are most common, if coherency can be obtained with the aid of appropriate oscillation sources and phase adjustment means even with micro waves and sound waves, Any of these may be used since the diffraction phenomenon utilized in the present invention may occur. However, as stated in the summary of this application, as a prerequisite, the separation distance between the first and second objects in the superimposition direction is such that the second diffraction grating is set so that the diffracted light exiting the first diffraction grating does not form a flat plate. It needs to be short enough to reach . As mentioned earlier, when the wave used is light, the distance between the two objects is several tens of tens of meters, as in the relationship between a semiconductor substrate and a lithography mask in X-ray lithography. This is on the order of μm, but in a long wavelength range, a large separation distance is naturally allowed.

Note that as a result, it is effective that the wave to be detected by the wave intensity detector is a first-order diffraction wave with high optical intensity. This also serves to increase the signal-to-noise ratio (S/N) for measurements.

〔Example〕

Embodiments of the present invention will be described in detail below. First, the embodiment shown in FIGS. 1 and 2 is considered to be one of the most basic embodiments of the present invention. be.

The description of the previous conventional example differs from the case in that the first object o1 is shown at the top for convenience, but in these first and second embodiments, the diffraction pattern provided on the first object o1 is In the grating D _a , two width regions W1 and W2 separated by a distance S in the in-plane direction are defined as first and second effective diffraction portions D _a-1 and D _a-2, respectively. It's summery.

There are several ways to specify this, as mentioned above in the section on effects, but first two means are shown in Figure 1, and the other means is shown in Figure 2. has been done.

One, which can be said in common to the first and second embodiments, is to locally irradiate coherent waves I _i and I _i only to the width regions W1 and W2, and the other , Similarly, as shown by the imaginary line in both FIGS. 1 and 2, a masking means MX is provided above the separated region S, and the incident wave I _i that is irradiated on the entire surface is also covered by this masking means.
The irradiation is blocked at a certain part of MX, so that first and second effective diffraction portions D _a-1 and D a-2 of widths W1 and W2 are formed substantially on both sides of the mask means MX when projected onto the _plane. The goal is to be able to do this.

In addition to or in place of these, a second
As shown in the figure, the first and second effective diffraction portions D _a-1 and D _a-2 are structurally completely separated regions, that is, regions in which diffraction grooves are cut separately and independently. It may be formed.

In any case, with respect to the first diffraction grating D _a that is separated in this way, the second object shown at the bottom in the figure in the first and second embodiments is
The diffraction grating D _b formed at o2 is a normal diffraction grating, and can be considered as a non-divided type diffraction grating in comparison with the divided type used in the present invention.
However, this second diffraction grating D _b is narrower than the first diffraction grating D _a and is smaller in width than the first diffraction grating when viewed from a plane projection (that is, when viewed through the top). A part that occupies the entire area on the second object that overlaps with the separated area S between the first and second effective diffraction parts D _a-1 and D _a-2 , or at least a part that includes the center in the in-plane direction. It is necessary that the structure be formed so as to occupy the area.

However, assuming that the grating periods d of both diffraction gratings D _a (D _a-1 , D _a-2 ) and D _b are the same for now, and coherent light of wavelength λ is used as the incident wave I _i , Either the coherent light I _i is locally irradiated onto the first diffraction grating D _a of the diffraction grating group under such geometrical arrangement, or the mask means MX
When the entire surface is irradiated with the aid of
A diffraction phenomenon occurs at D _a , and its first effective diffraction part
D _a-1 is the diffracted light diffracted in the diffraction angle + θ direction, preferably the first-order diffracted light _a-1 ⁺ , and the second effective diffraction part
D _a-2 is diffracted in the direction of diffraction angle -θ, which is also preferably first-order diffracted light _a-2 ^- , so that it enters the non-split diffraction grating D _b as the next and final stage as an effective incident component. Can be done. However, as a result, this can be done for the wave components that are effectively incident on the wave detectors DT and DT' (described later).
As is clear from the device of the present invention, the first and second effective diffraction portions D _a-1 ,
No means for selecting or extracting waves of a specific order using other optical components is provided between D _a-2 and the second diffraction grating D _b . On the other hand, as mentioned earlier, the first and second objects o1 and o2 are expected to be close to each other on the order of several tens of micrometers, and therefore, they will be incident on the second diffraction grating D _b . First, second effective diffraction part
The detected waves from D _a-1 and D _a-2 cannot become plane waves, so even if some kind of extremely fine optical system could be inserted between these two objects, it would be possible to detect a specific order. It is extremely difficult to define.

Even under such circumstances, according to the present invention, the two diffracted lights _a-1 ⁺ and _a-2 ^- focused on in the figure form interference fringes with a period of d/2 on the next-stage diffraction grating D _b . is further diffracted by this diffraction grating D _b to form output light s. In other words, in the case of the present invention, the diffracted lights _a-1 ⁺ , _a-2 ^-
The above interference fringes can be formed regardless of whether or not the wave is a plane wave.

The output light s obtained in this way can be converted into an electric signal and extracted by a suitable photodetector DT, but its intensity is determined by the interference fringes and the diffraction grating.
The relative displacement x with respect to D _b is proportional to cos ² (2πx/d), and each time the displacement x changes by d/2, it shows a maximum or minimum value.

Therefore, if we count the number of occurrences of the local maximum or minimum value, at least the object o1 at resolution d/2,
It is possible to roughly measure the relative displacement between o2,
As mentioned above, since the function form is extremely simple, by decomposing it using an appropriate threshold value, it is possible to perform measurements with considerably fine resolution.
The fact that a simple functional form can be obtained means that in the configuration according to the present invention as shown in FIGS. 1 and 2, even if the distance z between objects o1 and o2 is measured, _the This is because the position of the interference fringes formed in the image does not change.

In particular, even if first-order or higher-order reflection diffraction occurs at the second diffraction grating D _b and is returned to the first diffraction grating, the second diffraction grating
By appropriately setting the width Wo of D _b within a narrower range than the first diffraction grating D _a , it is possible to easily prevent the reflected and diffracted light component from being incident thereon. However, if the width of the second diffraction grating is too narrow, the intensity of the output diffracted light will drop significantly, and the allowable range of change in the separation distance in the superimposition direction between objects will become narrower, so it is difficult to make trade-offs between these factors in practice. This may need to be taken into consideration when designing. Furthermore, in terms of the probability of the reflected diffracted light being mixed into the output light, the embodiment shown in FIG. 2 is more advantageous than the embodiment shown in FIG. 1. This is because there is no room for reflection diffraction to occur in the separated region S between the first and second effective diffraction portions D _a-1 and D _a-2 .

Furthermore, in the above description, for the sake of explanation, the output light is
As shown in both the figure _{and FIG . s} ′.

That is, as shown by the virtual line detector DT', the diffraction grating portion provided in the spaced region S between the first and second effective diffraction portions of the first diffraction grating D _a is subjected to zero-order transmission diffraction. Waves (in this case, light) are output from the part of the first object o1 that has passed through the separated region S (Fig. 1) or the part of the first object o1 that has passed through the separated region S (Fig. 2).
limited to capturing. In other words, the detector DT shown by the solid line in the figure may have many uses where it cannot actually be placed there. second object
When o2 is a semiconductor substrate or a semiconductor wafer, its bottom surface is often supported by a holder or the like, and in that case, there is no room for providing a detector DT. On the other hand, if the reflected detection light is captured by the second diffraction grating, the light source of the incident light I _i to the first diffraction grating and the detector DT' can be placed on the same side, and the device configuration can be extremely precise. It becomes easier. Moreover, in the case of the present invention, in principle the detector
Since the number of DT′ may be just one, it can be made compact.

However, in the case of the embodiment shown in FIG. 1, the output light I _s ' due to reflection diffraction is diffracted there when passing through the spaced apart part S which should be the ineffective diffraction part of the first diffraction grating. As a result, there is a risk that the light may enter the detector DT′, and in such a case, the detector
It may be necessary to separate the output light by inserting a lens system (not shown) between the detector DT' and the first object on the same side as DT';
In the case of the embodiment shown in FIG. 2, there is no diffraction grating at least in the part S between them, so there is no problem of diffraction there, which is advantageous to that extent, and reduces the reduction in the intensity of the output light. It works well in that it doesn't cause problems, and the signal-to-noise ratio (S/N) continues to improve.

Here, we will discuss an actual example created by the present applicant, mainly due to the configuration shown in FIG.
d=0.8μm, S=75μm, incident light wavelength λ=0.6328μ
m, even if the distance z between the first and second objects o1 and o2 changes within a slightly larger range of 30 to 70 μm, the relative displacement between the first and second objects will be 0.01 μm.
We succeeded in making stable measurements with an accuracy of .

This value is extremely satisfactory, and is sufficiently applicable to the X-ray lithography technology as briefly mentioned above.

By the way, in the above, the pitch d of the first and second diffraction gratings D _a and D _b are both the same, but
If the lattice constant d of the second diffraction grating D _b is intentionally made different from the lattice constant d of the first diffraction grating or the first and second effective diffraction parts D _a-1 and D _a-2 , the output light can be It is also possible to make the emission direction different from the direction of the incident light. FIG. 3 shows such an embodiment. The basic configuration is exactly the same as in the first and second illustrated embodiments, but for simplicity, the first and second objects are
o2, mask means MX, detectors DT, DT', etc. are omitted.

In this embodiment, e.g. the first diffraction grating
Although the lattice constants d of _the first and second effective diffraction portions D _a-1 and D _a-2 of D a are the same, the second diffraction grating
When the lattice constant d of D _b is made different from those, the incident light I _i that entered the first and second effective diffraction parts D _a-1 and D _a-2 is diffracted, for example, in the first order by each of the diffraction parts, As shown in the figure, the output lights _{I s and} I _s ′, which are incident on the second diffraction grating D _b as a - and _a ⁺ and interfere with each other there ^, are further diffracted and output.
A specific angular relationship ±α with respect to the normal direction may be applied.

In this way, the component that the incident light I _i superimposes on the output lights I _s and I _s ' can be separated from the output light, and therefore, when the entire surface is irradiated with the incident light I _i , , the masking means MX, which was thought to be necessary in the previous embodiments, can be omitted.

For example, in the example created by the present applicant, the grating period d of the effective diffraction portions D _a-1 and D _a-2 is 0.8 μm, the grating period d of the grating D _b is 1.2 μm, and the wavelength λ is 0.6328 μm.
As a result of setting the distance S between the effective diffraction parts to 75 μm, even when the entire surface is irradiated with the incident light _I , the output light is well separated in the direction of ±α = 15.3° with respect to the normal direction.
I was able to obtain I _s and I _s ′. However, in the present invention, for the reason stated above, the output light to be detected is reflected by the second diffraction grating and passes through the separated region S between the first and second effective diffraction portions of the first diffraction grating. There is only the output light I _s ' that comes.

Furthermore, even if the distance between objects changes, it only changes within a small range, or if this incident light component can be removed in an external measurement system,
The above-mentioned masking means may be omitted in some cases even if no particular consideration is given to the lattice constant.

Incidentally, all of the embodiments described above were one-dimensional measurements along only the in-plane direction x.

However, the method of the present invention can be easily extended to in-plane two-dimensional measurements. Figure 4 shows a conceptual diagram of such a case.

In this case, the first object shown above in the diagram
o1 includes an x-direction diffraction grating D ax consisting of a pair of effective diffraction parts D _ax-1 and D _ax-2 for measuring relative displacement in the x-direction, and a diffraction grating D _ax that is orthogonal to these in the plane. y consisting of a pair of effective diffraction parts D _ay-1 and D _ay-2
A diffraction grating D _ay for directional relative displacement measurement is provided,
The second object o2 is provided with a diffraction grating D _a which is consequently formed on a grid so as to have periodicity in the x and y directions. Both effective diffraction parts of each diffraction grating D _ax-1 , D _ax-2 ; D _ay-1 , D _ay-2
The separation distance between them corresponds to the distance S mentioned above.

With this configuration, one effective diffraction part
between D _ax-1 , D _ax-2 and the periodic part in the x direction of the diffraction grid D _d , and between the effective diffraction parts D _ay-1 , D _{ay-2 in} the direction and the periodic part in the y direction of the diffraction grid D _d . The above-described exchange relationship of diffraction waves occurs independently between the two, and therefore, by measuring the intensity of the output light I _s ', it is possible to measure the amount of relative displacement in both the x and y directions. Of course, other practical considerations such as the configuration of the diffraction grating in each of the x and y directions, which have been described in the previous embodiments, can be directly applied to this embodiment.

Note that the position where each diffraction grating is provided relative to the object is not specified, and even in the case of Fig. 4, it is shown as being large in the center for convenience of explanation, but in general, You can also place it on the edge of the object. Furthermore, if the target object is not transparent to the waves or light used for measurement, the portion under the diffraction grating may be made transparent or empty to form a "passing window."

However, in general, it is expected that the method of the present invention will be preferred as a highly accurate measurement method for measuring minute displacements in the semiconductor technology field, such as alignment of substrates and masks in X-ray lithography, and therefore In this case, since the target object whose relative displacement is to be measured is often transparent, there is no need to intentionally form a wave transmission window.
Furthermore, as already mentioned, the coherent waves used in the present invention are not limited to the coherent light in the illustrated embodiment, but can also be obtained by electromagnetic waves such as sound waves and micro waves.

〔Effect of the invention〕

According to the present invention, the in-plane relative displacement between two objects superimposed in extremely close proximity can be calculated in an extremely small displacement area without being greatly influenced by the component that depends on the distance change in the direction of superposition. Can be measured with high precision.

Moreover, the configuration of the apparatus system required to carry out the method of the present invention is extremely simple, and the degree of freedom in design can be achieved in principle. It is practical enough.

For these reasons, there are almost no obstacles to realizing the present invention.

In particular, as seen in the examples above,
Since it can exhibit a resolution on the order of 0.01 μm, it has sufficient margin to meet future applications that require manufacturing accuracy on the order of sub-microns, such as semiconductor electronic circuit technology and Josefson integrated circuit technology. The practical value is extremely high.

[Brief explanation of drawings]

FIGS. 1 and 2 are explanatory diagrams of a basic embodiment of the relative displacement measurement device of the present invention, and FIG. 3 is an illustration of a basic embodiment of the relative displacement measuring device of the present invention, and FIG. An explanatory diagram of an embodiment of the present invention. Figure 4 is an explanatory diagram of an embodiment of the present invention that can measure the amount of relative displacement in two-dimensional directions in a plane. FIG. 2 is an explanatory diagram of a conventional method for measuring the amount of relative displacement between objects. In the figure, o1 and o2 are objects, D _a and D _b are diffraction gratings, and D _d
is a grid-like diffraction grating having periodic grooves in both x and y directions, D _a-1 and D _a-2 are effective diffraction areas, S is the separation distance between the effective diffraction areas, and I _i
is the incident light, I _s , I _s ′ are the output lights, and θ, α are the particular and single diffraction angles of interest output from each grating.

Claims (1)

[Claims] 1. When a first diffraction grating is irradiated with a coherent wave from one side, the first diffraction grating is diffracted by the first diffraction grating, and the first diffraction grating is irradiated with a coherent wave from the other side. The wave passing through the surface is made to directly enter the second diffraction grating, which is superimposed close to the other surface so that the wave does not become a plane wave, without going through an optical system. The intensity of the wave diffracted by the diffraction grating is detected by a detector, and based on the intensity change of the detected wave, the first object provided with the first diffraction grating and the second diffraction grating are detected. A device for measuring the amount of relative displacement in an in-plane direction orthogonal to the superimposition direction between second objects provided with; The second diffraction grating is narrower than the first diffraction grating. All or at least the center in the in-plane direction of the region on the second object that overlaps with the spaced region between the first and second effective diffraction portions of the first diffraction grating when viewed from a plane projection. after being diffracted and reflected by the second diffraction grating, the wave intensity detector occupies a portion including the first and second effective diffraction portions of the first diffraction grating. A relative displacement amount measuring device characterized in that it is provided at a position to receive waves outputted through the above-mentioned separated region between the two. 2. The first and second effective diffraction portions of the first diffraction grating are independent area portions that are structurally separated from each other on the first object. Relative displacement amount measuring device as described in section.